Fresenius J Anal Chem (1998) 362 : 270–273

© Springer-Verlag 1998

O R I G I N A L PA P E R

Norbert Binding · Holger Kläning · Uwe Karst · Wilhelm Pötter · Peter A. Czeschinski · Ute Witting

Analytical reliability of carbonyl compound determination using 1,5-dansylhydrazine-derivatization

Received: 8 January 1998 / Revised: 9 March 1998 / Accepted: 15 March 1998

Abstract Solid sorbents coated with the fluorescent reagent 5-dimethylaminonaphthalene-1-sulfohydrazide (dansylhydrazine, DNSH) have been used for derivatization and quantitative determination of airborne carbonyl compounds, for example in investigations on atmospheric pollution. The evaluation of this derivatization reaction presented here revealed that, for several reasons, it may not be recommended when sampling is performed with impingers containing liquid reagent solutions. Derivative yields came out to be strongly influenced by water and phosphoric acid which are essential for sufficient derivatization rates, but also responsible for the degradation of derivatives. Even at water and acid concentrations considered as an optimal compromise between accelerating and degrading effects, the analytical reliability of the method can only be guaranteed under controlled laboratory conditions. The reduced or lacking reactivity of DNSH towards aromatic aldehydes or aliphatic and aromatic ketones is an additional finding discarding the DNSH method for routine air monitoring at least when impingers are used for sampling.

of analytical methods for airborne carbonyl compounds have been developed [5–9], most of them with a range of determination suited for monitoring work place and indoor air, but not for low environmental atmospheric concentrations which play an important role in atmospheric chemistry [10, 11]. Thus, the need for more sensitive analytical methods is obvious. One of the most promising is the DNSH-method [12, 13]. Dansylhydrazine (5-dimethylaminonaphthalene-1-sulfohydrazide, DNSH) is used for the acid catalyzed derivatization of aldehydes and ketones to the corresponding hydrazones which can be sensitively and selectively detected by fluorescence. After several applications of reagent-coated solid sorbent tubes to monitor atmospheric concentrations of carbonyl compounds [4, 10, 11], instabilities of the hydrazones due to water interferences have been reported [14]. These interferences could be almost entirely eliminated by using octadecyl silica as sorbent in combination with trichloroacetic acid [14]. Furthermore, reduced or lacking reactivity of DNSH to derivatize ketones has been described [10]. In this paper, investigations on the reactivity of DNSH in liquid phase and the stability of the corresponding carbonyl hydrazones are presented.

Introduction Carbonyl compounds, especially formaldehyde, are frequently used in industrial as well as in non-industrial applications [1]. The versatile applicability and the volatility of low molecular weight compounds unavoidably lead to contaminations of the atmosphere and thus to occupational and environmental exposure [2–4]. Quite a number

Experimental

N. Binding (Y) · H. Kläning · P.A. Czeschinski · U. Witting Institut für Arbeitsmedizin der Universität Münster, Robert-Koch-Strasse 51, D-48 149 Münster, Germany e-mail: [email protected]

Reagents. Analytical grade 5-dimethylaminonaphthalene-1-sulfohydrazide and 1-dimethylaminonaphthalene-5-sulfonic acid were from Aldrich, Deisenhofen, FRG; phosphoric acid (85%), formaldehyde (37%), acetaldehyde, propanal, butanal, pentanal, hexanal, benzaldehyde, acetone, 2-butanone, 2-pentanone, 2-hexanone, cyclohexanone and acetophenone all were of analytical grade, and acetonitrile was HPLC grade from Merck, Darmstadt, FRG. Derivatization of carbonyl compounds for spectrofluorophotometry:

U. Karst · W. Pötter Anorganisch-Chemisches Institut der Universität Münster, Lehrstuhl für Analytische Chemie, Wilhelm-Klemm-Strasse 8, D-48 149 Münster, Germany

Apparatus. A gradient high performance liquid chromatograph was used with the following components: 2 solvent delivery modules 114, organizer 340, autosampler Promis with 20 µL loop, System Gold Software (Beckman Instruments) and Fluorescence HPLC Monitor RF 530 (Shimadzu). Separations were done on a C18 reverse phase column (LiChrospher 100-RP-18, 5 µm, 250 × 4 mm, Merck, Darmstadt, FRG). Fluorescence spectra were obtained with a Spectrofluorophotometer RF-5301 PC (Shimadzu).

271 As hydrazones could not be synthesized without considerable decomposition during purification procedures, solutions of the derivatives were prepared by adding 3 mL of a solution of the respective carbonyl compound in acetonitrile (37.5 mmol/L, 50fold excess) to 3 mL of a solution of DNSH in acetonitrile (0.75 mmol/L). The reaction mixture was left to stay for 1 day and after control for complete consumption of DNSH by HPLC used for spectrofluorophotometry. Quenching of fluorescence by remaining free carbonyl compound could be excluded by control measurements with mixtures containing the carbonyl compound in a 200fold excess. Influence of water and phosphoric acid on the fluorescence properties of DNSH. Solutions of DNSH in acetonitrile (10 mg/mL) containing water (0 to 80%) or phosphoric acid (0 to 1%) were used for spectrofluorophotometry. Spectrofluorophotometry. After dilution (1 : 10) of the hydrazone and DNSH samples, the optimum excitation and emission wavelengths for each compound were determined by scanning the wavelength range between 250 and 700 nm at a slit of 3 nm. Investigations on the influence of water and phosphoric acid on reaction rates and derivative stability. 1 mL of a solution of DNSH in acetonitrile (10 mg/mL) and 1 mL formaldehyde in acetonitrile (0.5 mg/mL) were added to 18 mL acetonitrile solutions with water and phosphoric acid concentrations adjusted to result in 0; 1; 5; 10; 50% water and for each water concentration in 0; 0.1; 0.2; 0.5; 1; 2% phosphoric acid in the final volume of 20 mL. The reaction mixtures were analyzed after 1, 6, 11, 16, 21, 31, 46, 61, 91, and 121 min and then every 24 h for seven days by HPLC. For acetaldehyde, benzaldehyde, acetone, and acetophenone the same method was used, but only with a water content of 5% and a phosphoric acid concentration of 0.1%.

351–359 nm) and emission wavelengths (λem = 522– 527 nm) both differ by less than 10 nm (Table 1). Excitation and emission intensities as well lie in a very narrow range (cf. Table 1). With a Stokes-shift of about 170 nm fluorescence reabsorption should be neglectable. Based on these results, identical excitation and emission wavelengths can be chosen for HPLC fluorescence detection of hydrazones and DNSH. Furthermore, as excitation and emission intensities are nearly identical, calibration for one hydrazone can be used for all others. The investigations on the influence of water and phosphoric acid on the fluorescence properties of DNSH in acetonitrile revealed that an increasing water content results in an increasing shift of excitation and emission wavelengths to smaller values (Fig. 1). This shift may be of considerable influence on reproducibility when gradient separation with water-containing solvent mixtures is used, but not for isocratic separation conditions. Phosphoric acid leads to strong quenching effects in spectrofluorophotometry, but not in HPLC detection, since phosphoric acid is eluted from the column with the solvent front. According to Fig. 2, water and phosphoric acid are also involved in the derivatization reaction, since hydrazone

HPLC conditions. Separation was performed with an isocratic mixture of acetonitrile and water (50/50 v/v) and a flow of 1.5 mL/min; fluorescence detection wavelengths were λex = 355 nm, λem = 525 nm.

Results and discussion The fluorescence spectra of several DNS-hydrazones and of DNSH show that excitation wavelengths (λex = Table 1 Wavelengths and intensities of excitation and emission for selected DNS-hydrazones DNS-hydrazone

Formaldehyde Acetaldehyde Acetone Propanal Butanal 2-Butanone Pentanal 2-Pentanone Cyclohexanone Hexanal 2-Hexanone Benzaldehyde Acetophenone

Excitation

Fig. 1 Water-induced shift of excitation and emission wavelengths of 1,5-dansylhydrazine (––––: 0% water, – – – 80% water)

Emission

Wavelength [nm]

Intensity [scaled]

Wavelength [nm]

Intensity [scaled]

356 353 355 352 351 357 354 355 358 353 351 357 359

397 393 395 399 401 394 397 396 391 398 400 396 395

527 526 523 526 525 526 522 524 527 525 523 527 522

388 385 389 386 388 384 387 389 381 383 386 384 389

Fig. 2 Reaction scheme of derivatization and degradation reactions involved in formaldehyde-DNS-hydrazone formation

272

Fig. 3 Influence of water concentration on reaction rates of formaldehyde DNS-hydrazone formation and degradation

too, showed a considerable reduction of reaction rate, while acetophenone, representing an aromatic ketone, did not react at all. Compared to the frequently used 2,4-dinitrophenylhydrazine (DNPH) method for the determination of carbonyl compounds [5, 7], the DNSH method has the convincing advantage of a by decades higher sensitivity. Thus, at the first sight, it should be the method of choice, especially for the determination of low environmental carbonyl concentrations and has already been used by several groups [4, 10, 11] to investigate atmospheric contaminations. Nevertheless, the re-evaluation of the DNSH method presented here raises some reasonable doubts about its analytical reliability and practicability, especially when liquid phase samplers are used.

Conclusions

Fig. 4 Influence of phosphoric acid concentration on reaction rates of formaldehyde DNS-hydrazone formation and degradation

formation is acid catalyzed [12], and water may contribute to degradation reactions of reagent and derivatives [14]. Especially the water content of acid-free reaction solutions is of considerable influence on hydrazone formation rates as shown in Fig. 3 for formaldehyde. Only water concentrations of 50% or more result in an increased, but still unsatisfactory derivatization rate during the first two hours, but already lead to pronounced degradation of the formaldehyde hydrazone during the following seven days. For investigations on the influence of phosphoric acid on the derivatization of formaldehyde, solutions containing 5% water were used, since lower water contents led to precipitation of the hydrazone. The accelerating effect of phosphoric acid on hydrazone formation rates is shown in Fig. 4. Hydrazone formation is complete after 30 to 40 min for H3PO4 concentrations of 0.1 or 0.2%, but already after 1 min for 2.0%. But with an acid content of 2.0%, the degradation reaction was accelerated, too. From these experiments it was concluded that water and phosphoric acid contents must be in a narrow concentration range for optimum reaction conditions. 5% water and 0.1% phosphoric acid were considered as the optimal compromise between accelerating and degrading effects. Derivatization of other carbonyl compounds was performed under these optimized conditions. Compared to formaldehyde, acetaldehyde reacts with a similar rate, while benzaldehyde derivatization is rather slow. Acetone,

The reaction of carbonyl compounds with DNSH is determined by a delicate balance between derivatization and degradation reactions which both are accelerated with increasing water and phosphoric acid contents of reaction solutions. Thus, a strict control of water and acid concentrations throughout the analytical procedures is a basic requirement to assure reproducible results. This can only be guaranteed under laboratory conditions, but not for ambient air samples where humidity might increase the water content. Using impingers, solvent evaporation during sampling will result in an uncontrollable increase of water and acid concentrations and thus in irreproducible yields of derivatives. Furthermore, due to the time-dependent degradation of derivatives, the analytical result is determined by the time passed by between sampling and HPLC analysis. Storage of samples until analysis which, in most cases, is unavoidable in routine occupational and environmental air monitoring, thus is prohibited. With the additional findings of reduced or lacking reactivity of DNSH for aromatic aldehydes and aliphatic and aromatic ketones, the DNSH method, though more sensitive than for example the DNPH method, must be classified as less versatile and cannot be recommended for routine air monitoring of carbonyl compounds.

References 1. Reuss G, Disteldorf W, Grundler O, Hilt A (1988) In: Gerhartz W, Yamamoto YS, Elvers B, Rounsaville JF, Schulz, G (eds) Ullmann’s Encyclopedia of Industrial Chemistry, Vol. A11, VCH Publishers, New York, 5th rev. edn, pp 619–651 2. Binding N, Witting U (1990) Int Arch Occup Environ Health 62 : 233–238 3. Sexton K, Petreas MX, Liu K-S (1989) Environ Sci Technol 23 : 985–988 4. Nondek L, Rodier DR, Birks JW (1992) Environ Sci Technol 26, 1174–1178 5. Binding N, Thiewens S, Witting U (1986) Staub – Reinhalt Luft 46 : 444–446 6. NIOSH (1994) Ketones I, method 1300, issue 2. NIOSH Manual of Analytical Methods, 4th edn 1994

273 7. Kuwata K, Uebori M, Yamasaki YJ (1979) Chromatogr Sci 17 : 264–268 8. Lehmkuhl DW, Birks JW (1996) J Chromatogr A 740 : 71–78 9. Leibrock E, Slemr J (1997) Atmospheric Environment 31 : 3329–3339 10. Schmied W, Przewosnik M, Bächmann K (1989) Fresenius Z Anal Chem 335 : 464–468 11. Nondek L, Milofsky RE, Birks JW (1991) Chromatographia 32 : 33–39

12. Chayen R, Dvir R, Gould S, Harell A (1971) Anal Biochem 42 : 283–286 13. Bächmann K, Haag I, Han KY, Schmitzer RQ (1993) Fresenius J Anal Chem 346 : 786–788 14. Rodier DR, Nondek L, Birks JW (1993) Environ Sci Technol 27 : 2814–2820